By using an acoustic replay device such as a loud speaker or a head phone, a stereophonic acoustic technique to localize a sound image (as a virtual sound source) at an arbitrary (frontward and rearward, leftward and rightward) position of a listener is well known.
As to a sound localization apparatus of conventional stereophonic acoustic technique, a head-related transfer function (from a desired position to localize the sound image to both ears of the listener) is convoluted with an audio signal, and the audio signal is presented to the listener. As a result, the sound image can be localized at the desired position.
In this sound localization apparatus used for the acoustic replay apparatus, realization of a function to adjust an emphasis degree of feeling of localization (to be presented to the listener) based on the listener's liking is desired.
However, in order to adjust the emphasis degree of feeling of localization for the listener, it is insufficient that a sound pressure at the listener's ears (when a sound source really exists) is accurately reappeared by using the head-related transfer function. In localization processing of the sound image based on the head-related transfer function, a factor to affect on the emphasis degree of feeling of localization is not clear, and the emphasis degree of feeling of localization of the sound image is difficult to be adjusted.